you-should-have-a-fave-data-story

A data visualisation is just a graphical representation of data.

Dataviz are not necessarily intelligible - which is where data stories come in.

A data visualisation is just a graphical representation of data.

Dataviz are not necessarily intelligible - and can be quite confusing.

Data storytelling is about communicating. Data stories make dataviz meaningful and engaging.

I’ll demonstrate what I mean by sharing my favourite data story - “A Day in the Life of Americans” by Nathan Yau.

Let me take you through some of the comments I’ve heard people make about this data story.

❤️ There are lots of bright, clear colours.

🛏️ No matter what time of day it is, there are always people sleeping.

🤔 There are fewer people working, and more people doing housework than you might expect for the early afternoon.

🚌 It’s really clever to put travelling as grey and in the middle.

Commuting is a liminal state.
You’re likely not in a traffic jam because you want to be.

😵‍💫 But wait a minute…

What does it mean when there’s a different coloured dot in a cluster?
Is there really someone sleeping while at work - are they using a nap pod?

🧐 Actually, how sure are we that this is exactly what people are doing at 4:20pm?

Who are these people?

Where is the data from?

🤯 This data visualisation doesn’t contain exactly what people are doing minute by minute.

The actual data from the American Time Use Survey is too complicated to understand in its raw form.

This is why I love Nathan Yau’s data visualisation.

It transforms this complex, messy, human data into something we can begin to understand and explore.

This is beautiful data storytelling.

This is my 2nd favourite data story.

It’s from an open-access paper on food pairing in regional cuisines.

It’s very much the opposite of the first data story

Nathan Yau’s “Day in the Life” was very much an enjoyable choose your own adventure data story.

This is very much a difficult read as bibliophiles might say. It’s dense.

Lines between ingredients indicate they share flavour chemicals.

Colour indicates ingredient type.

Size indicates prelevance in the 56,000 recipes studied in the paper it comes from.

That’s fascinating!

The research paper investigates which regional cuisines prefer their recipes to contain similar or disimilar flavours!

I love that this has been studied and that this visualisation tries to bring all that information together.

But. There are big data quality issues here.

Are “brown rice” and “rice” definitely two distinct ingredients?

Liver comes in three types - beef, chicken and undefined?

It’s a little unusual to have such a flawed data story as my 2nd favourite.
But it has personal importance to me.

The data behind the chart was the first dataset I ever (struggled to) read into R.

It’s important when you choose your own favourite data stories that they resonate with you.

Just before I help you find your own favourite chart - here’s my 3rd favourite data story.

It’s by Caitlin Hudson and she first shared it on Twitter back in 2020.

I can’t find any fault with this data story. It’s understated beauty.

However. The maternity leave policies in the US are the worst amongst the world’s richest countries.

There are lots of faults to find with that.